Motion estimation compatible with multiple standards

Abstract

A motion estimation engine may be implemented to support multiple video encoding standards. The motion estimation engine may be designed to support two macroblock partitioning modes: one for frame type video signals and the other for mixed frame-field type video signals. Additionally, the motion estimation engine provides the mixing unidirectional option (forward/backward) and the mixing bidirectional option. Furthermore, the motion estimation engine may use a unified 5-tap interpolation filter for fractional macroblock search during motion estimation.

What is claimed is: 
     
       1. An apparatus comprising:
 a processor to determine whether a video standard allows each sub-block to have its own decision whether prediction is forward or backward, if not, finding the best macroblock partition in each direction and choosing the better of the two directions, if so, merging forward and backward motion vectors for all sub-blocks and finding the best overall motion block partition, determine whether the video standard allows each sub-block to have its own decision on whether prediction is unidirectional or bidirectional, if not, comparing the bidirectional overall result with all sub-blocks being unidirectional and bidirectional, and if so, comparing bidirectional results for each sub-block; and 
 a memory coupled to said processor. 
 
     
     
       2. The apparatus of  claim 1 , said processor to select the best sub-block motion vectors and then determine whether a video standard allows each sub-block to have its own decision whether prediction is forward or backward. 
     
     
       3. The apparatus of  claim 1 , said processor to choose the better of two directions by choosing the direction with smaller distortion. 
     
     
       4. The apparatus of  claim 1 , said processor to perform fractional refinement on a chosen partition. 
     
     
       5. The apparatus of  claim 1 , said processor to perform an intra-frame search in parallel with an inter-frame search. 
     
     
       6. A method comprising:
 determining whether a video standard allows each sub-block to have its own decision whether prediction is forward or backward; 
 if not, finding the best macroblock partition in each direction and choosing the better of the two directions; 
 if so, merging forward and backward motion vectors for all sub-blocks and finding the best overall motion block partition; 
 determining whether the video standard allows each sub-block to have its own decision on whether prediction is unidirectional or bidirectional; 
 if not, comparing the bidirectional overall result with all sub-blocks being unidirectional and bidirectional; and 
 if so, comparing bidirectional results for each sub-block. 
 
     
     
       7. The method of  claim 6  including selecting the best sub-block motion vectors and then determining whether a video standard allows each sub-block to have its own decision whether prediction is forward or backward. 
     
     
       8. The method of  claim 6  wherein choosing the better of two directions includes choosing the direction with smaller distortion. 
     
     
       9. The method of  claim 6  including performing fractional refinement on a chosen partition. 
     
     
       10. The method of  claim 6  including performing an intra-frame search in parallel with an inter-frame search. 
     
     
       11. One or more non-transitory machine-readable storage media storing instructions, that when executed by a processing platform, cause said processing platform to perform operations including:
 determining whether a video standard allows each sub-block to have its own decision whether prediction is forward or backward; 
 if not, finding the best macroblock partition in each direction and choosing the better of the two directions; 
 if so, merging forward and backward motion vectors for all sub-blocks and finding the best overall motion block partition; 
 determining whether the video standard allows each sub-block to have its own decision on whether prediction is unidirectional or bidirectional; 
 if not, comparing the bidirectional overall result with all sub-blocks being unidirectional and bidirectional; and 
 if so, comparing bidirectional results for each sub-block. 
 
     
     
       12. The media of  claim 11 , said operations including selecting the best sub-block motion vectors and then determining whether a video standard allows each sub-block to have its own decision whether prediction is forward or backward. 
     
     
       13. The media of  claim 11 , wherein choosing the better of two directions includes choosing the direction with smaller distortion. 
     
     
       14. The media of  claim 11 , said operations including performing fractional refinement on a chosen partition. 
     
     
       15. The media of  claim 11 , said operations including performing an intra-frame search in parallel with an inter-frame search.